PORTALE DELLA DIDATTICA

Ricerca CERCA
  KEYWORD

IEIIT/CNR COMPUTER ENGINEERING AND NETWORKS GROUP

Low-cost temperature estimation based on machine learning techniques and Raspberry Pi

keywords ARTIFICIAL NEURAL NETWORKS, HVAC SYSTEMS, MACHINE LEARNING, REGRESSION TECHNIQUES, TEMPERATURE SENSORS

Reference persons STEFANO SCANZIO

Research Groups IEIIT/CNR COMPUTER ENGINEERING AND NETWORKS GROUP

Thesis type APPLIED RESEARCH, RESEARCH

Description The thesis focuses on the use of machine learning (ML) algorithms to estimate, in a low-cost way, the temperature in indoor environments. The thesis work consists of two main steps, that is the development of (1) a data acquisition system and (2) a temperature estimation system. For step (1), both PCs and Raspberry Pi devices will be employed. Raspberry Pis will be used for acquiring temperature sensor data while PCs will be employed to measure features related/influencing the temperature (e.g., the temperature measured by sensors in CPU, motherboard, hard disk, etc., as well as CPU usage, interrupts rate, etc.).
The final goal of the thesis work is to use the simple data acquired by the PCs (and the other features) to estimate room temperature through ML regression techniques especially based on artificial neural networks.

See also  https://www.skenz.it/ss/theses

Required skills Basic knowledge of the Linux operating system, good programming skills (especially Python programming language), basic knowledge of machine learning.

Notes Much of the knowledge can be acquired during the thesis.


Deadline 28/10/2021      PROPONI LA TUA CANDIDATURA




© Politecnico di Torino
Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY
Contatti